Research on Recommendation Algorithms Based on Cloud Models in Probabilistic Linguistic Environments

To solve the problem that the traditional cloud model can't directly process the textual review information in the recommendation algorithm, this paper combines the merits of the cloud model in transforming qualitative and quantitative knowledge with the multi-granularity advantages of probabilistic linguistic term sets in representing uncertain information, and proposes a recommendation algorithm based on cloud model in probabilistic language environment. Initially, this paper quantifies the attributes in the review text based on the probabilistic linguistic term set. Subsequently, the maximum deviation method is used to determine the weight of each attribute in the evaluation information of the product to be recommended, and the comprehensive evaluation number and attribute weight are converted into the digital characteristic value of the cloud model by using the backward cloud generator. Finally, the products are recommended and sorted based on the digital characteristic value of the cloud model. The algorithm is applied to the recommendation of 10 hotels, and the results show that the method is effective and practical, enriching the application of cloud models in the recommendation field.